{
  "architectures": [
    "NystromformerModel"
  ],
  "attention_probs_dropout_prob": 0.1,
  "bos_token_id": 0,
  "conv_kernel_size": 65,
  "eos_token_id": 2,
  "hidden_act": "gelu_new",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "inv_coeff_init_option": false,
  "layer_norm_eps": 1e-05,
  "max_position_embeddings": 510,
  "model_type": "nystromformer",
  "num_attention_heads": 12,
  "num_hidden_layers": 12,
  "num_landmarks": 64,
  "pad_token_id": 1,
  "quantization_config": {
    "_load_in_4bit": true,
    "_load_in_8bit": false,
    "bnb_4bit_compute_dtype": "bfloat16",
    "bnb_4bit_quant_storage": "uint8",
    "bnb_4bit_quant_type": "nf4",
    "bnb_4bit_use_double_quant": false,
    "llm_int8_enable_fp32_cpu_offload": false,
    "llm_int8_has_fp16_weight": false,
    "llm_int8_skip_modules": null,
    "llm_int8_threshold": 6.0,
    "load_in_4bit": true,
    "load_in_8bit": false,
    "quant_method": "bitsandbytes"
  },
  "segment_means_seq_len": 64,
  "tokenizer_class": "AlbertTokenizer",
  "torch_dtype": "float32",
  "transformers_version": "4.53.1",
  "type_vocab_size": 2,
  "use_cache": true,
  "vocab_size": 30000
}